pgsql/doc/src/sgml/html/runtime-config-client.html - master - Ao/MaxReport

pgsql/doc/src/sgml/html/runtime-config-client.html
887 строк69 KB
AoAnima
first_commit
13 июл 2026, 17:47
13 июл 2026, 17:47f1a670a
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>19.11. Client Connection Defaults</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="runtime-config-vacuum.html" title="19.10. Vacuuming" /><link rel="next" href="runtime-config-locks.html" title="19.12. Lock Management" /></head><body id="docContent" class="container-fluid col-10"><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">19.11. Client Connection Defaults</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="runtime-config-vacuum.html" title="19.10. Vacuuming">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="runtime-config.html" title="Chapter 19. Server Configuration">Up</a></td><th width="60%" align="center">Chapter 19. Server Configuration</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 18.4 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="runtime-config-locks.html" title="19.12. Lock Management">Next</a></td></tr></table><hr /></div><div class="sect1" id="RUNTIME-CONFIG-CLIENT"><div class="titlepage"><div><div><h2 class="title" style="clear: both">19.11. Client Connection Defaults <a href="#RUNTIME-CONFIG-CLIENT" class="id_link">#</a></h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="runtime-config-client.html#RUNTIME-CONFIG-CLIENT-STATEMENT">19.11.1. Statement Behavior</a></span></dt><dt><span class="sect2"><a href="runtime-config-client.html#RUNTIME-CONFIG-CLIENT-FORMAT">19.11.2. Locale and Formatting</a></span></dt><dt><span class="sect2"><a href="runtime-config-client.html#RUNTIME-CONFIG-CLIENT-PRELOAD">19.11.3. Shared Library Preloading</a></span></dt><dt><span class="sect2"><a href="runtime-config-client.html#RUNTIME-CONFIG-CLIENT-OTHER">19.11.4. Other Defaults</a></span></dt></dl></div><div class="sect2" id="RUNTIME-CONFIG-CLIENT-STATEMENT"><div class="titlepage"><div><div><h3 class="title">19.11.1. Statement Behavior <a href="#RUNTIME-CONFIG-CLIENT-STATEMENT" class="id_link">#</a></h3></div></div></div><div class="variablelist"><dl class="variablelist"><dt id="GUC-CLIENT-MIN-MESSAGES"><span class="term"><code class="varname">client_min_messages</code> (<code class="type">enum</code>)
      <a id="id-1.6.6.14.2.2.1.1.3" class="indexterm"></a>
      </span> <a href="#GUC-CLIENT-MIN-MESSAGES" class="id_link">#</a></dt><dd><p>
        Controls which
        <a class="link" href="runtime-config-logging.html#RUNTIME-CONFIG-SEVERITY-LEVELS" title="Table 19.2. Message Severity Levels">message levels</a>
        are sent to the client.
        Valid values are <code class="literal">DEBUG5</code>,
        <code class="literal">DEBUG4</code>, <code class="literal">DEBUG3</code>, <code class="literal">DEBUG2</code>,
        <code class="literal">DEBUG1</code>, <code class="literal">LOG</code>, <code class="literal">NOTICE</code>,
        <code class="literal">WARNING</code>, and <code class="literal">ERROR</code>.
        Each level includes all the levels that follow it.  The later the level,
        the fewer messages are sent.  The default is
        <code class="literal">NOTICE</code>.  Note that <code class="literal">LOG</code> has a different
        rank here than in <a class="xref" href="runtime-config-logging.html#GUC-LOG-MIN-MESSAGES">log_min_messages</a>.
       </p><p>
        <code class="literal">INFO</code> level messages are always sent to the client.
       </p></dd><dt id="GUC-SEARCH-PATH"><span class="term"><code class="varname">search_path</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.2.2.2.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.2.2.2.1.4" class="indexterm"></a>
      </span> <a href="#GUC-SEARCH-PATH" class="id_link">#</a></dt><dd><p>
        This variable specifies the order in which schemas are searched
        when an object (table, data type, function, etc.) is referenced by a
        simple name with no schema specified.  When there are objects of
        identical names in different schemas, the one found first
        in the search path is used.  An object that is not in any of the
        schemas in the search path can only be referenced by specifying
        its containing schema with a qualified (dotted) name.
       </p><p>
        The value for <code class="varname">search_path</code> must be a comma-separated
        list of schema names.  Any name that is not an existing schema, or is
        a schema for which the user does not have <code class="literal">USAGE</code>
        permission, is silently ignored.
       </p><p>
        If one of the list items is the special name
        <code class="literal">$user</code>, then the schema having the name returned by
        <code class="function">CURRENT_USER</code> is substituted, if there is such a schema
        and the user has <code class="literal">USAGE</code> permission for it.
        (If not, <code class="literal">$user</code> is ignored.)
       </p><p>
        The system catalog schema, <code class="literal">pg_catalog</code>, is always
        searched, whether it is mentioned in the path or not.  If it is
        mentioned in the path then it will be searched in the specified
        order.  If <code class="literal">pg_catalog</code> is not in the path then it will
        be searched <span class="emphasis"><em>before</em></span> searching any of the path items.
       </p><p>
        Likewise, the current session's temporary-table schema,
        <code class="literal">pg_temp_<em class="replaceable"><code>nnn</code></em></code>, is always searched if it
        exists.  It can be explicitly listed in the path by using the
        alias <code class="literal">pg_temp</code><a id="id-1.6.6.14.2.2.2.2.5.3" class="indexterm"></a>.  If it is not listed in the path then
        it is searched first (even before <code class="literal">pg_catalog</code>).  However,
        the temporary schema is only searched for relation (table, view,
        sequence, etc.) and data type names.  It is never searched for
        function or operator names.
       </p><p>
        When objects are created without specifying a particular target
        schema, they will be placed in the first valid schema named in
        <code class="varname">search_path</code>.  An error is reported if the search
        path is empty.
       </p><p>
        The default value for this parameter is
        <code class="literal">"$user", public</code>.
        This setting supports shared use of a database (where no users
        have private schemas, and all share use of <code class="literal">public</code>),
        private per-user schemas, and combinations of these.  Other
        effects can be obtained by altering the default search path
        setting, either globally or per-user.
       </p><p>
        For more information on schema handling, see
        <a class="xref" href="ddl-schemas.html" title="5.10. Schemas">Section 5.10</a>.  In particular, the default
        configuration is suitable only when the database has a single user or
        a few mutually-trusting users.
       </p><p>
        The current effective value of the search path can be examined
        via the <acronym class="acronym">SQL</acronym> function
        <code class="function">current_schemas</code>
        (see <a class="xref" href="functions-info.html" title="9.27. System Information Functions and Operators">Section 9.27</a>).
        This is not quite the same as
        examining the value of <code class="varname">search_path</code>, since
        <code class="function">current_schemas</code> shows how the items
        appearing in <code class="varname">search_path</code> were resolved.
       </p></dd><dt id="GUC-ROW-SECURITY"><span class="term"><code class="varname">row_security</code> (<code class="type">boolean</code>)
      <a id="id-1.6.6.14.2.2.3.1.3" class="indexterm"></a>
      </span> <a href="#GUC-ROW-SECURITY" class="id_link">#</a></dt><dd><p>
        This variable controls whether to raise an error in lieu of applying a
        row security policy.  When set to <code class="literal">on</code>, policies apply
        normally.  When set to <code class="literal">off</code>, queries fail which would
        otherwise apply at least one policy.  The default is <code class="literal">on</code>.
        Change to <code class="literal">off</code> where limited row visibility could cause
        incorrect results; for example, <span class="application">pg_dump</span> makes that
        change by default.  This variable has no effect on roles which bypass
        every row security policy, to wit, superusers and roles with
        the <code class="literal">BYPASSRLS</code> attribute.
       </p><p>
        For more information on row security policies,
        see <a class="xref" href="sql-createpolicy.html" title="CREATE POLICY"><span class="refentrytitle">CREATE POLICY</span></a>.
       </p></dd><dt id="GUC-DEFAULT-TABLE-ACCESS-METHOD"><span class="term"><code class="varname">default_table_access_method</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.2.2.4.1.3" class="indexterm"></a>
      </span> <a href="#GUC-DEFAULT-TABLE-ACCESS-METHOD" class="id_link">#</a></dt><dd><p>
        This parameter specifies the default table access method to use when
        creating tables or materialized views if the <code class="command">CREATE</code>
        command does not explicitly specify an access method, or when
        <code class="command">SELECT ... INTO</code> is used, which does not allow
        specifying a table access method. The default is <code class="literal">heap</code>.
       </p></dd><dt id="GUC-DEFAULT-TABLESPACE"><span class="term"><code class="varname">default_tablespace</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.2.2.5.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.2.2.5.1.4" class="indexterm"></a>
      </span> <a href="#GUC-DEFAULT-TABLESPACE" class="id_link">#</a></dt><dd><p>
        This variable specifies the default tablespace in which to create
        objects (tables and indexes) when a <code class="command">CREATE</code> command does
        not explicitly specify a tablespace.
       </p><p>
        The value is either the name of a tablespace, or an empty string
        to specify using the default tablespace of the current database.
        If the value does not match the name of any existing tablespace,
        <span class="productname">PostgreSQL</span> will automatically use the default
        tablespace of the current database.  If a nondefault tablespace
        is specified, the user must have <code class="literal">CREATE</code> privilege
        for it, or creation attempts will fail.
       </p><p>
        This variable is not used for temporary tables; for them,
        <a class="xref" href="runtime-config-client.html#GUC-TEMP-TABLESPACES">temp_tablespaces</a> is consulted instead.
       </p><p>
        This variable is also not used when creating databases.
        By default, a new database inherits its tablespace setting from
        the template database it is copied from.
       </p><p>
        If this parameter is set to a value other than the empty string
        when a partitioned table is created, the partitioned table's
        tablespace will be set to that value, which will be used as
        the default tablespace for partitions created in the future,
        even if <code class="varname">default_tablespace</code> has changed since then.
       </p><p>
        For more information on tablespaces,
        see <a class="xref" href="manage-ag-tablespaces.html" title="22.6. Tablespaces">Section 22.6</a>.
       </p></dd><dt id="GUC-DEFAULT-TOAST-COMPRESSION"><span class="term"><code class="varname">default_toast_compression</code> (<code class="type">enum</code>)
      <a id="id-1.6.6.14.2.2.6.1.3" class="indexterm"></a>
      </span> <a href="#GUC-DEFAULT-TOAST-COMPRESSION" class="id_link">#</a></dt><dd><p>
        This variable sets the default
        <a class="link" href="storage-toast.html" title="66.2. TOAST">TOAST</a>
        compression method for values of compressible columns.
        (This can be overridden for individual columns by setting
        the <code class="literal">COMPRESSION</code> column option in
        <code class="command">CREATE TABLE</code> or
        <code class="command">ALTER TABLE</code>.)
        The supported compression methods are <code class="literal">pglz</code> and
        (if <span class="productname">PostgreSQL</span> was compiled with
        <code class="option">--with-lz4</code>) <code class="literal">lz4</code>.
        The default is <code class="literal">pglz</code>.
       </p></dd><dt id="GUC-TEMP-TABLESPACES"><span class="term"><code class="varname">temp_tablespaces</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.2.2.7.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.2.2.7.1.4" class="indexterm"></a>
      </span> <a href="#GUC-TEMP-TABLESPACES" class="id_link">#</a></dt><dd><p>
        This variable specifies tablespaces in which to create temporary
        objects (temp tables and indexes on temp tables) when a
        <code class="command">CREATE</code> command does not explicitly specify a tablespace.
        Temporary files for purposes such as sorting large data sets
        are also created in these tablespaces.
       </p><p>
        The value is a list of names of tablespaces.  When there is more than
        one name in the list, <span class="productname">PostgreSQL</span> chooses a random
        member of the list each time a temporary object is to be created;
        except that within a transaction, successively created temporary
        objects are placed in successive tablespaces from the list.
        If the selected element of the list is an empty string,
        <span class="productname">PostgreSQL</span> will automatically use the default
        tablespace of the current database instead.
       </p><p>
        When <code class="varname">temp_tablespaces</code> is set interactively, specifying a
        nonexistent tablespace is an error, as is specifying a tablespace for
        which the user does not have <code class="literal">CREATE</code> privilege.  However,
        when using a previously set value, nonexistent tablespaces are
        ignored, as are tablespaces for which the user lacks
        <code class="literal">CREATE</code> privilege.  In particular, this rule applies when
        using a value set in <code class="filename">postgresql.conf</code>.
       </p><p>
        The default value is an empty string, which results in all temporary
        objects being created in the default tablespace of the current
        database.
       </p><p>
        See also <a class="xref" href="runtime-config-client.html#GUC-DEFAULT-TABLESPACE">default_tablespace</a>.
       </p></dd><dt id="GUC-CHECK-FUNCTION-BODIES"><span class="term"><code class="varname">check_function_bodies</code> (<code class="type">boolean</code>)
      <a id="id-1.6.6.14.2.2.8.1.3" class="indexterm"></a>
      </span> <a href="#GUC-CHECK-FUNCTION-BODIES" class="id_link">#</a></dt><dd><p>
        This parameter is normally on. When set to <code class="literal">off</code>, it
        disables validation of the routine body string during <a class="xref" href="sql-createfunction.html" title="CREATE FUNCTION"><span class="refentrytitle">CREATE FUNCTION</span></a> and <a class="xref" href="sql-createprocedure.html" title="CREATE PROCEDURE"><span class="refentrytitle">CREATE PROCEDURE</span></a>.  Disabling validation avoids side
        effects of the validation process, in particular preventing false
        positives due to problems such as forward references.
        Set this parameter
        to <code class="literal">off</code> before loading functions on behalf of other
        users; <span class="application">pg_dump</span> does so automatically.
       </p></dd><dt id="GUC-DEFAULT-TRANSACTION-ISOLATION"><span class="term"><code class="varname">default_transaction_isolation</code> (<code class="type">enum</code>)
      <a id="id-1.6.6.14.2.2.9.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.2.2.9.1.4" class="indexterm"></a>
      </span> <a href="#GUC-DEFAULT-TRANSACTION-ISOLATION" class="id_link">#</a></dt><dd><p>
        Each SQL transaction has an isolation level, which can be
        either <span class="quote">“<span class="quote">read uncommitted</span>”</span>, <span class="quote">“<span class="quote">read
        committed</span>”</span>, <span class="quote">“<span class="quote">repeatable read</span>”</span>, or
        <span class="quote">“<span class="quote">serializable</span>”</span>.  This parameter controls the
        default isolation level of each new transaction. The default
        is <span class="quote">“<span class="quote">read committed</span>”</span>.
       </p><p>
        Consult <a class="xref" href="mvcc.html" title="Chapter 13. Concurrency Control">Chapter 13</a> and <a class="xref" href="sql-set-transaction.html" title="SET TRANSACTION"><span class="refentrytitle">SET TRANSACTION</span></a> for more information.
       </p></dd><dt id="GUC-DEFAULT-TRANSACTION-READ-ONLY"><span class="term"><code class="varname">default_transaction_read_only</code> (<code class="type">boolean</code>)
      <a id="id-1.6.6.14.2.2.10.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.2.2.10.1.4" class="indexterm"></a>
      </span> <a href="#GUC-DEFAULT-TRANSACTION-READ-ONLY" class="id_link">#</a></dt><dd><p>
        A read-only SQL transaction cannot alter non-temporary tables.
        This parameter controls the default read-only status of each new
        transaction. The default is <code class="literal">off</code> (read/write).
       </p><p>
        Consult <a class="xref" href="sql-set-transaction.html" title="SET TRANSACTION"><span class="refentrytitle">SET TRANSACTION</span></a> for more information.
       </p></dd><dt id="GUC-DEFAULT-TRANSACTION-DEFERRABLE"><span class="term"><code class="varname">default_transaction_deferrable</code> (<code class="type">boolean</code>)
      <a id="id-1.6.6.14.2.2.11.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.2.2.11.1.4" class="indexterm"></a>
      </span> <a href="#GUC-DEFAULT-TRANSACTION-DEFERRABLE" class="id_link">#</a></dt><dd><p>
        When running at the <code class="literal">serializable</code> isolation level,
        a deferrable read-only SQL transaction may be delayed before
        it is allowed to proceed.  However, once it begins executing
        it does not incur any of the overhead required to ensure
        serializability; so serialization code will have no reason to
        force it to abort because of concurrent updates, making this
        option suitable for long-running read-only transactions.
        </p><p>
        This parameter controls the default deferrable status of each
        new transaction.  It currently has no effect on read-write
        transactions or those operating at isolation levels lower
        than <code class="literal">serializable</code>. The default is <code class="literal">off</code>.
       </p><p>
        Consult <a class="xref" href="sql-set-transaction.html" title="SET TRANSACTION"><span class="refentrytitle">SET TRANSACTION</span></a> for more information.
       </p></dd><dt id="GUC-TRANSACTION-ISOLATION"><span class="term"><code class="varname">transaction_isolation</code> (<code class="type">enum</code>)
      <a id="id-1.6.6.14.2.2.12.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.2.2.12.1.4" class="indexterm"></a>
      </span> <a href="#GUC-TRANSACTION-ISOLATION" class="id_link">#</a></dt><dd><p>
        This parameter reflects the current transaction's isolation level.
        At the beginning of each transaction, it is set to the current value
        of <a class="xref" href="runtime-config-client.html#GUC-DEFAULT-TRANSACTION-ISOLATION">default_transaction_isolation</a>.
        Any subsequent attempt to change it is equivalent to a <a class="xref" href="sql-set-transaction.html" title="SET TRANSACTION"><span class="refentrytitle">SET TRANSACTION</span></a> command.
       </p></dd><dt id="GUC-TRANSACTION-READ-ONLY"><span class="term"><code class="varname">transaction_read_only</code> (<code class="type">boolean</code>)
      <a id="id-1.6.6.14.2.2.13.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.2.2.13.1.4" class="indexterm"></a>
      </span> <a href="#GUC-TRANSACTION-READ-ONLY" class="id_link">#</a></dt><dd><p>
        This parameter reflects the current transaction's read-only status.
        At the beginning of each transaction, it is set to the current value
        of <a class="xref" href="runtime-config-client.html#GUC-DEFAULT-TRANSACTION-READ-ONLY">default_transaction_read_only</a>.
        Any subsequent attempt to change it is equivalent to a <a class="xref" href="sql-set-transaction.html" title="SET TRANSACTION"><span class="refentrytitle">SET TRANSACTION</span></a> command.
       </p></dd><dt id="GUC-TRANSACTION-DEFERRABLE"><span class="term"><code class="varname">transaction_deferrable</code> (<code class="type">boolean</code>)
      <a id="id-1.6.6.14.2.2.14.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.2.2.14.1.4" class="indexterm"></a>
      </span> <a href="#GUC-TRANSACTION-DEFERRABLE" class="id_link">#</a></dt><dd><p>
        This parameter reflects the current transaction's deferrability status.
        At the beginning of each transaction, it is set to the current value
        of <a class="xref" href="runtime-config-client.html#GUC-DEFAULT-TRANSACTION-DEFERRABLE">default_transaction_deferrable</a>.
        Any subsequent attempt to change it is equivalent to a <a class="xref" href="sql-set-transaction.html" title="SET TRANSACTION"><span class="refentrytitle">SET TRANSACTION</span></a> command.
       </p></dd><dt id="GUC-SESSION-REPLICATION-ROLE"><span class="term"><code class="varname">session_replication_role</code> (<code class="type">enum</code>)
      <a id="id-1.6.6.14.2.2.15.1.3" class="indexterm"></a>
      </span> <a href="#GUC-SESSION-REPLICATION-ROLE" class="id_link">#</a></dt><dd><p>
        Controls firing of replication-related triggers and rules for the
        current session.
        Possible values are <code class="literal">origin</code> (the default),
        <code class="literal">replica</code> and <code class="literal">local</code>.
        Setting this parameter results in discarding any previously cached
        query plans.
        Only superusers and users with the appropriate <code class="literal">SET</code>
        privilege can change this setting.
       </p><p>
        The intended use of this setting is that logical replication systems
        set it to <code class="literal">replica</code> when they are applying replicated
        changes.  The effect of that will be that triggers and rules (that
        have not been altered from their default configuration) will not fire
        on the replica.  See the <a class="link" href="sql-altertable.html" title="ALTER TABLE"><code class="command">ALTER TABLE</code></a> clauses
        <code class="literal">ENABLE TRIGGER</code> and <code class="literal">ENABLE RULE</code>
        for more information.
       </p><p>
        PostgreSQL treats the settings <code class="literal">origin</code> and
        <code class="literal">local</code> the same internally.  Third-party replication
        systems may use these two values for their internal purposes, for
        example using <code class="literal">local</code> to designate a session whose
        changes should not be replicated.
       </p><p>
        Since foreign keys are implemented as triggers, setting this parameter
        to <code class="literal">replica</code> also disables all foreign key checks,
        which can leave data in an inconsistent state if improperly used.
       </p></dd><dt id="GUC-STATEMENT-TIMEOUT"><span class="term"><code class="varname">statement_timeout</code> (<code class="type">integer</code>)
      <a id="id-1.6.6.14.2.2.16.1.3" class="indexterm"></a>
      </span> <a href="#GUC-STATEMENT-TIMEOUT" class="id_link">#</a></dt><dd><p>
        Abort any statement that takes more than the specified amount of time.
        If <code class="varname">log_min_error_statement</code> is set
        to <code class="literal">ERROR</code> or lower, the statement that timed out
        will also be logged.
        If this value is specified without units, it is taken as milliseconds.
        A value of zero (the default) disables the timeout.
       </p><p>
        The timeout is measured from the time a command arrives at the
        server until it is completed by the server.  If multiple SQL
        statements appear in a single simple-query message, the timeout
        is applied to each statement separately.
        (<span class="productname">PostgreSQL</span> versions before 13 usually
        treated the timeout as applying to the whole query string.)
        In extended query protocol, the timeout starts running when any
        query-related message (Parse, Bind, Execute, Describe) arrives, and
        it is canceled by completion of an Execute or Sync message.
       </p><p>
        Setting <code class="varname">statement_timeout</code> in
        <code class="filename">postgresql.conf</code> is not recommended because it would
        affect all sessions.
       </p></dd><dt id="GUC-TRANSACTION-TIMEOUT"><span class="term"><code class="varname">transaction_timeout</code> (<code class="type">integer</code>)
      <a id="id-1.6.6.14.2.2.17.1.3" class="indexterm"></a>
      </span> <a href="#GUC-TRANSACTION-TIMEOUT" class="id_link">#</a></dt><dd><p>
        Terminate any session that spans longer than the specified amount of
        time in a transaction. The limit applies both to explicit transactions
        (started with <code class="command">BEGIN</code>) and to an implicitly started
        transaction corresponding to a single statement.
        If this value is specified without units, it is taken as milliseconds.
        A value of zero (the default) disables the timeout.
       </p><p>
        If <code class="varname">transaction_timeout</code> is shorter or equal to
        <code class="varname">idle_in_transaction_session_timeout</code> or <code class="varname">statement_timeout</code>
        then the longer timeout is ignored.
       </p><p>
        Setting <code class="varname">transaction_timeout</code> in
        <code class="filename">postgresql.conf</code> is not recommended because it would
        affect all sessions.
       </p><div class="note"><h3 class="title">Note</h3><p>
         Prepared transactions are not subject to this timeout.
        </p></div></dd><dt id="GUC-LOCK-TIMEOUT"><span class="term"><code class="varname">lock_timeout</code> (<code class="type">integer</code>)
      <a id="id-1.6.6.14.2.2.18.1.3" class="indexterm"></a>
      </span> <a href="#GUC-LOCK-TIMEOUT" class="id_link">#</a></dt><dd><p>
        Abort any statement that waits longer than the specified amount of
        time while attempting to acquire a lock on a table, index,
        row, or other database object.  The time limit applies separately to
        each lock acquisition attempt.  The limit applies both to explicit
        locking requests (such as <code class="command">LOCK TABLE</code>, or <code class="command">SELECT
        FOR UPDATE</code> without <code class="literal">NOWAIT</code>) and to implicitly-acquired
        locks.
        If this value is specified without units, it is taken as milliseconds.
        A value of zero (the default) disables the timeout.
       </p><p>
        Unlike <code class="varname">statement_timeout</code>, this timeout can only occur
        while waiting for locks.  Note that if <code class="varname">statement_timeout</code>
        is nonzero, it is rather pointless to set <code class="varname">lock_timeout</code> to
        the same or larger value, since the statement timeout would always
        trigger first.  If <code class="varname">log_min_error_statement</code> is set to
        <code class="literal">ERROR</code> or lower, the statement that timed out will be
        logged.
       </p><p>
        Setting <code class="varname">lock_timeout</code> in
        <code class="filename">postgresql.conf</code> is not recommended because it would
        affect all sessions.
       </p></dd><dt id="GUC-IDLE-IN-TRANSACTION-SESSION-TIMEOUT"><span class="term"><code class="varname">idle_in_transaction_session_timeout</code> (<code class="type">integer</code>)
      <a id="id-1.6.6.14.2.2.19.1.3" class="indexterm"></a>
      </span> <a href="#GUC-IDLE-IN-TRANSACTION-SESSION-TIMEOUT" class="id_link">#</a></dt><dd><p>
        Terminate any session that has been idle (that is, waiting for a
        client query) within an open transaction for longer than the
        specified amount of time.
        If this value is specified without units, it is taken as milliseconds.
        A value of zero (the default) disables the timeout.
       </p><p>
        This option can be used to ensure that idle sessions do not hold
        locks for an unreasonable amount of time.  Even when no significant
        locks are held, an open transaction prevents vacuuming away
        recently-dead tuples that may be visible only to this transaction;
        so remaining idle for a long time can contribute to table bloat.
        See <a class="xref" href="routine-vacuuming.html" title="24.1. Routine Vacuuming">Section 24.1</a> for more details.
       </p></dd><dt id="GUC-IDLE-SESSION-TIMEOUT"><span class="term"><code class="varname">idle_session_timeout</code> (<code class="type">integer</code>)
      <a id="id-1.6.6.14.2.2.20.1.3" class="indexterm"></a>
      </span> <a href="#GUC-IDLE-SESSION-TIMEOUT" class="id_link">#</a></dt><dd><p>
        Terminate any session that has been idle (that is, waiting for a
        client query), but not within an open transaction, for longer than
        the specified amount of time.
        If this value is specified without units, it is taken as milliseconds.
        A value of zero (the default) disables the timeout.
       </p><p>
        Unlike the case with an open transaction, an idle session without a
        transaction imposes no large costs on the server, so there is less
        need to enable this timeout
        than <code class="varname">idle_in_transaction_session_timeout</code>.
       </p><p>
        Be wary of enforcing this timeout on connections made through
        connection-pooling software or other middleware, as such a layer
        may not react well to unexpected connection closure.  It may be
        helpful to enable this timeout only for interactive sessions,
        perhaps by applying it only to particular users.
       </p></dd><dt id="GUC-BYTEA-OUTPUT"><span class="term"><code class="varname">bytea_output</code> (<code class="type">enum</code>)
      <a id="id-1.6.6.14.2.2.21.1.3" class="indexterm"></a>
      </span> <a href="#GUC-BYTEA-OUTPUT" class="id_link">#</a></dt><dd><p>
        Sets the output format for values of type <code class="type">bytea</code>.
        Valid values are <code class="literal">hex</code> (the default)
        and <code class="literal">escape</code> (the traditional PostgreSQL
        format).  See <a class="xref" href="datatype-binary.html" title="8.4. Binary Data Types">Section 8.4</a> for more
        information.  The <code class="type">bytea</code> type always
        accepts both formats on input, regardless of this setting.
       </p></dd><dt id="GUC-XMLBINARY"><span class="term"><code class="varname">xmlbinary</code> (<code class="type">enum</code>)
      <a id="id-1.6.6.14.2.2.22.1.3" class="indexterm"></a>
      </span> <a href="#GUC-XMLBINARY" class="id_link">#</a></dt><dd><p>
        Sets how binary values are to be encoded in XML.  This applies
        for example when <code class="type">bytea</code> values are converted to
        XML by the functions <code class="function">xmlelement</code> or
        <code class="function">xmlforest</code>.  Possible values are
        <code class="literal">base64</code> and <code class="literal">hex</code>, which
        are both defined in the XML Schema standard.  The default is
        <code class="literal">base64</code>.  For further information about
        XML-related functions, see <a class="xref" href="functions-xml.html" title="9.15. XML Functions">Section 9.15</a>.
       </p><p>
        The actual choice here is mostly a matter of taste,
        constrained only by possible restrictions in client
        applications.  Both methods support all possible values,
        although the hex encoding will be somewhat larger than the
        base64 encoding.
       </p></dd><dt id="GUC-XMLOPTION"><span class="term"><code class="varname">xmloption</code> (<code class="type">enum</code>)
      <a id="id-1.6.6.14.2.2.23.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.2.2.23.1.4" class="indexterm"></a>
      <a id="id-1.6.6.14.2.2.23.1.5" class="indexterm"></a>
      </span> <a href="#GUC-XMLOPTION" class="id_link">#</a></dt><dd><p>
        Sets whether <code class="literal">DOCUMENT</code> or
        <code class="literal">CONTENT</code> is implicit when converting between
        XML and character string values.  See <a class="xref" href="datatype-xml.html" title="8.13. XML Type">Section 8.13</a> for a description of this.  Valid
        values are <code class="literal">DOCUMENT</code> and
        <code class="literal">CONTENT</code>.  The default is
        <code class="literal">CONTENT</code>.
       </p><p>
        According to the SQL standard, the command to set this option is
</p><pre class="synopsis">
SET XML OPTION { DOCUMENT | CONTENT };
</pre><p>
        This syntax is also available in PostgreSQL.
       </p></dd><dt id="GUC-GIN-PENDING-LIST-LIMIT"><span class="term"><code class="varname">gin_pending_list_limit</code> (<code class="type">integer</code>)
      <a id="id-1.6.6.14.2.2.24.1.3" class="indexterm"></a>
      </span> <a href="#GUC-GIN-PENDING-LIST-LIMIT" class="id_link">#</a></dt><dd><p>
        Sets the maximum size of a GIN index's pending list, which is used
        when <code class="literal">fastupdate</code> is enabled. If the list grows
        larger than this maximum size, it is cleaned up by moving
        the entries in it to the index's main GIN data structure in bulk.
        If this value is specified without units, it is taken as kilobytes.
        The default is four megabytes (<code class="literal">4MB</code>). This setting
        can be overridden for individual GIN indexes by changing
        index storage parameters.
         See <a class="xref" href="gin.html#GIN-FAST-UPDATE" title="65.4.4.1. GIN Fast Update Technique">Section 65.4.4.1</a> and <a class="xref" href="gin.html#GIN-TIPS" title="65.4.5. GIN Tips and Tricks">Section 65.4.5</a>
         for more information.
       </p></dd><dt id="GUC-CREATEROLE-SELF-GRANT"><span class="term"><code class="varname">createrole_self_grant</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.2.2.25.1.3" class="indexterm"></a>
      </span> <a href="#GUC-CREATEROLE-SELF-GRANT" class="id_link">#</a></dt><dd><p>
        If a user who has <code class="literal">CREATEROLE</code> but not
        <code class="literal">SUPERUSER</code> creates a role, and if this
        is set to a non-empty value, the newly-created role will be granted
        to the creating user with the options specified. The value must be
        <code class="literal">set</code>, <code class="literal">inherit</code>, or a
        comma-separated list of these. The default value is an empty string,
        which disables the feature.
       </p><p>
        The purpose of this option is to allow a <code class="literal">CREATEROLE</code>
        user who is not a superuser to automatically inherit, or automatically
        gain the ability to <code class="literal">SET ROLE</code> to, any created users.
        Since a <code class="literal">CREATEROLE</code> user is always implicitly granted
        <code class="literal">ADMIN OPTION</code> on created roles, that user could
        always execute a <code class="literal">GRANT</code> statement that would achieve
        the same effect as this setting. However, it can be convenient for
        usability reasons if the grant happens automatically. A superuser
        automatically inherits the privileges of every role and can always
        <code class="literal">SET ROLE</code> to any role, and this setting can be used
        to produce a similar behavior for <code class="literal">CREATEROLE</code> users
        for users which they create.
       </p></dd><dt id="GUC-EVENT-TRIGGERS"><span class="term"><code class="varname">event_triggers</code> (<code class="type">boolean</code>)
      <a id="id-1.6.6.14.2.2.26.1.3" class="indexterm"></a>
     </span> <a href="#GUC-EVENT-TRIGGERS" class="id_link">#</a></dt><dd><p>
       Allow temporarily disabling execution of event triggers in order to
       troubleshoot and repair faulty event triggers. All event triggers will
       be disabled by setting it to <code class="literal">false</code>. Setting the value
       to <code class="literal">true</code> allows all event triggers to fire, this
       is the default value. Only superusers and users with the appropriate
       <code class="literal">SET</code> privilege can change this setting.
      </p></dd><dt id="GUC-RESTRICT-NONSYSTEM-RELATION-KIND"><span class="term"><code class="varname">restrict_nonsystem_relation_kind</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.2.2.27.1.3" class="indexterm"></a>
     </span> <a href="#GUC-RESTRICT-NONSYSTEM-RELATION-KIND" class="id_link">#</a></dt><dd><p>
       Set relation kinds for which access to non-system relations is prohibited.
       The value takes the form of a comma-separated list of relation kinds.
       Currently, the supported relation kinds are <code class="literal">view</code> and
       <code class="literal">foreign-table</code>.
      </p></dd></dl></div></div><div class="sect2" id="RUNTIME-CONFIG-CLIENT-FORMAT"><div class="titlepage"><div><div><h3 class="title">19.11.2. Locale and Formatting <a href="#RUNTIME-CONFIG-CLIENT-FORMAT" class="id_link">#</a></h3></div></div></div><div class="variablelist"><dl class="variablelist"><dt id="GUC-DATESTYLE"><span class="term"><code class="varname">DateStyle</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.3.2.1.1.3" class="indexterm"></a>
      </span> <a href="#GUC-DATESTYLE" class="id_link">#</a></dt><dd><p>
        Sets the display format for date and time values, as well as the
        rules for interpreting ambiguous date input values. For
        historical reasons, this variable contains two independent
        components: the output format specification (<code class="literal">ISO</code>,
        <code class="literal">Postgres</code>, <code class="literal">SQL</code>, or <code class="literal">German</code>)
        and the input/output specification for year/month/day ordering
        (<code class="literal">DMY</code>, <code class="literal">MDY</code>, or <code class="literal">YMD</code>). These
        can be set separately or together. The keywords <code class="literal">Euro</code>
        and <code class="literal">European</code> are synonyms for <code class="literal">DMY</code>; the
        keywords <code class="literal">US</code>, <code class="literal">NonEuro</code>, and
        <code class="literal">NonEuropean</code> are synonyms for <code class="literal">MDY</code>. See
        <a class="xref" href="datatype-datetime.html" title="8.5. Date/Time Types">Section 8.5</a> for more information. The
        built-in default is <code class="literal">ISO, MDY</code>, but
        <span class="application">initdb</span> will initialize the
        configuration file with a setting that corresponds to the
        behavior of the chosen <code class="varname">lc_time</code> locale.
       </p></dd><dt id="GUC-INTERVALSTYLE"><span class="term"><code class="varname">IntervalStyle</code> (<code class="type">enum</code>)
      <a id="id-1.6.6.14.3.2.2.1.3" class="indexterm"></a>
      </span> <a href="#GUC-INTERVALSTYLE" class="id_link">#</a></dt><dd><p>
        Sets the display format for interval values.
        The value <code class="literal">sql_standard</code> will produce
        output matching <acronym class="acronym">SQL</acronym> standard interval literals.
        The value <code class="literal">postgres</code> (which is the default) will produce
        output matching <span class="productname">PostgreSQL</span> releases prior to 8.4
        when the <a class="xref" href="runtime-config-client.html#GUC-DATESTYLE">DateStyle</a>
        parameter was set to <code class="literal">ISO</code>.
        The value <code class="literal">postgres_verbose</code> will produce output
        matching <span class="productname">PostgreSQL</span> releases prior to 8.4
        when the <code class="varname">DateStyle</code>
        parameter was set to non-<code class="literal">ISO</code> output.
        The value <code class="literal">iso_8601</code> will produce output matching the time
        interval <span class="quote">“<span class="quote">format with designators</span>”</span> defined in section
        4.4.3.2 of ISO 8601.
       </p><p>
        The <code class="varname">IntervalStyle</code> parameter also affects the
        interpretation of ambiguous interval input.  See
        <a class="xref" href="datatype-datetime.html#DATATYPE-INTERVAL-INPUT" title="8.5.4. Interval Input">Section 8.5.4</a> for more information.
       </p></dd><dt id="GUC-TIMEZONE"><span class="term"><code class="varname">TimeZone</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.3.2.3.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.3.2.3.1.4" class="indexterm"></a>
      </span> <a href="#GUC-TIMEZONE" class="id_link">#</a></dt><dd><p>
        Sets the time zone for displaying and interpreting time stamps.
        The built-in default is <code class="literal">GMT</code>, but that is typically
        overridden in <code class="filename">postgresql.conf</code>; <span class="application">initdb</span>
        will install a setting there corresponding to its system environment.
        See <a class="xref" href="datatype-datetime.html#DATATYPE-TIMEZONES" title="8.5.3. Time Zones">Section 8.5.3</a> for more information.
       </p></dd><dt id="GUC-TIMEZONE-ABBREVIATIONS"><span class="term"><code class="varname">timezone_abbreviations</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.3.2.4.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.3.2.4.1.4" class="indexterm"></a>
      </span> <a href="#GUC-TIMEZONE-ABBREVIATIONS" class="id_link">#</a></dt><dd><p>
        Sets the collection of additional time zone abbreviations that
        will be accepted by the server for datetime input (beyond any
        abbreviations defined by the current <code class="varname">TimeZone</code>
        setting).  The default is <code class="literal">'Default'</code>,
        which is a collection that works in most of the world; there are
        also <code class="literal">'Australia'</code> and <code class="literal">'India'</code>,
        and other collections can be defined for a particular installation.
        See <a class="xref" href="datetime-config-files.html" title="B.4. Date/Time Configuration Files">Section B.4</a> for more information.
       </p></dd><dt id="GUC-EXTRA-FLOAT-DIGITS"><span class="term"><code class="varname">extra_float_digits</code> (<code class="type">integer</code>)
      <a id="id-1.6.6.14.3.2.5.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.3.2.5.1.4" class="indexterm"></a>
      <a id="id-1.6.6.14.3.2.5.1.5" class="indexterm"></a>
      </span> <a href="#GUC-EXTRA-FLOAT-DIGITS" class="id_link">#</a></dt><dd><p>
        This parameter adjusts the number of digits used for textual output of
        floating-point values, including <code class="type">float4</code>, <code class="type">float8</code>,
        and geometric data types.
       </p><p>
        If the value is 1 (the default) or above, float values are output in
        shortest-precise format; see <a class="xref" href="datatype-numeric.html#DATATYPE-FLOAT" title="8.1.3. Floating-Point Types">Section 8.1.3</a>. The
        actual number of digits generated depends only on the value being
        output, not on the value of this parameter. At most 17 digits are
        required for <code class="type">float8</code> values, and 9 for <code class="type">float4</code>
        values. This format is both fast and precise, preserving the original
        binary float value exactly when correctly read. For historical
        compatibility, values up to 3 are permitted.
       </p><p>
        If the value is zero or negative, then the output is rounded to a
        given decimal precision. The precision used is the standard number of
        digits for the type (<code class="literal">FLT_DIG</code>
        or <code class="literal">DBL_DIG</code> as appropriate) reduced according to the
        value of this parameter. (For example, specifying -1 will cause
        <code class="type">float4</code> values to be output rounded to 5 significant
        digits, and <code class="type">float8</code> values
        rounded to 14 digits.) This format is slower and does not preserve all
        the bits of the binary float value, but may be more human-readable.
       </p><div class="note"><h3 class="title">Note</h3><p>
         The meaning of this parameter, and its default value, changed
         in <span class="productname">PostgreSQL</span> 12;
         see <a class="xref" href="datatype-numeric.html#DATATYPE-FLOAT" title="8.1.3. Floating-Point Types">Section 8.1.3</a> for further discussion.
        </p></div></dd><dt id="GUC-CLIENT-ENCODING"><span class="term"><code class="varname">client_encoding</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.3.2.6.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.3.2.6.1.4" class="indexterm"></a>
      </span> <a href="#GUC-CLIENT-ENCODING" class="id_link">#</a></dt><dd><p>
        Sets the client-side encoding (character set).
        The default is to use the database encoding.
        The character sets supported by the <span class="productname">PostgreSQL</span>
        server are described in <a class="xref" href="multibyte.html#MULTIBYTE-CHARSET-SUPPORTED" title="23.3.1. Supported Character Sets">Section 23.3.1</a>.
       </p></dd><dt id="GUC-LC-MESSAGES"><span class="term"><code class="varname">lc_messages</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.3.2.7.1.3" class="indexterm"></a>
      </span> <a href="#GUC-LC-MESSAGES" class="id_link">#</a></dt><dd><p>
        Sets the language in which messages are displayed.  Acceptable
        values are system-dependent; see <a class="xref" href="locale.html" title="23.1. Locale Support">Section 23.1</a> for
        more information.  If this variable is set to the empty string
        (which is the default) then the value is inherited from the
        execution environment of the server in a system-dependent way.
       </p><p>
        On some systems, this locale category does not exist.  Setting
        this variable will still work, but there will be no effect.
        Also, there is a chance that no translated messages for the
        desired language exist.  In that case you will continue to see
        the English messages.
       </p><p>
        Only superusers and users with the appropriate <code class="literal">SET</code>
        privilege can change this setting.
       </p></dd><dt id="GUC-LC-MONETARY"><span class="term"><code class="varname">lc_monetary</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.3.2.8.1.3" class="indexterm"></a>
      </span> <a href="#GUC-LC-MONETARY" class="id_link">#</a></dt><dd><p>
        Sets the locale to use for formatting monetary amounts, for
        example with the <code class="function">to_char</code> family of
        functions.  Acceptable values are system-dependent; see <a class="xref" href="locale.html" title="23.1. Locale Support">Section 23.1</a> for more information.  If this variable is
        set to the empty string (which is the default) then the value
        is inherited from the execution environment of the server in a
        system-dependent way.
       </p></dd><dt id="GUC-LC-NUMERIC"><span class="term"><code class="varname">lc_numeric</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.3.2.9.1.3" class="indexterm"></a>
      </span> <a href="#GUC-LC-NUMERIC" class="id_link">#</a></dt><dd><p>
        Sets the locale to use for formatting numbers, for example
        with the <code class="function">to_char</code> family of
        functions. Acceptable values are system-dependent; see <a class="xref" href="locale.html" title="23.1. Locale Support">Section 23.1</a> for more information.  If this variable is
        set to the empty string (which is the default) then the value
        is inherited from the execution environment of the server in a
        system-dependent way.
       </p></dd><dt id="GUC-LC-TIME"><span class="term"><code class="varname">lc_time</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.3.2.10.1.3" class="indexterm"></a>
      </span> <a href="#GUC-LC-TIME" class="id_link">#</a></dt><dd><p>
        Sets the locale to use for formatting dates and times, for example
        with the <code class="function">to_char</code> family of
        functions. Acceptable values are system-dependent; see <a class="xref" href="locale.html" title="23.1. Locale Support">Section 23.1</a> for more information.  If this variable is
        set to the empty string (which is the default) then the value
        is inherited from the execution environment of the server in a
        system-dependent way.
       </p></dd><dt id="GUC-ICU-VALIDATION-LEVEL"><span class="term"><code class="varname">icu_validation_level</code> (<code class="type">enum</code>)
      <a id="id-1.6.6.14.3.2.11.1.3" class="indexterm"></a>
      </span> <a href="#GUC-ICU-VALIDATION-LEVEL" class="id_link">#</a></dt><dd><p>
        When ICU locale validation problems are encountered, controls which
        <a class="link" href="runtime-config-logging.html#RUNTIME-CONFIG-SEVERITY-LEVELS" title="Table 19.2. Message Severity Levels">message level</a> is
        used to report the problem. Valid values are
        <code class="literal">DISABLED</code>, <code class="literal">DEBUG5</code>,
        <code class="literal">DEBUG4</code>, <code class="literal">DEBUG3</code>,
        <code class="literal">DEBUG2</code>, <code class="literal">DEBUG1</code>,
        <code class="literal">INFO</code>, <code class="literal">NOTICE</code>,
        <code class="literal">WARNING</code>, <code class="literal">ERROR</code>, and
        <code class="literal">LOG</code>.
       </p><p>
        If set to <code class="literal">DISABLED</code>, does not report validation
        problems at all. Otherwise reports problems at the given message
        level. The default is <code class="literal">WARNING</code>.
       </p></dd><dt id="GUC-DEFAULT-TEXT-SEARCH-CONFIG"><span class="term"><code class="varname">default_text_search_config</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.3.2.12.1.3" class="indexterm"></a>
      </span> <a href="#GUC-DEFAULT-TEXT-SEARCH-CONFIG" class="id_link">#</a></dt><dd><p>
        Selects the text search configuration that is used by those variants
        of the text search functions that do not have an explicit argument
        specifying the configuration.
        See <a class="xref" href="textsearch.html" title="Chapter 12. Full Text Search">Chapter 12</a> for further information.
        The built-in default is <code class="literal">pg_catalog.simple</code>, but
        <span class="application">initdb</span> will initialize the
        configuration file with a setting that corresponds to the
        chosen <code class="varname">lc_ctype</code> locale, if a configuration
        matching that locale can be identified.
       </p></dd></dl></div></div><div class="sect2" id="RUNTIME-CONFIG-CLIENT-PRELOAD"><div class="titlepage"><div><div><h3 class="title">19.11.3. Shared Library Preloading <a href="#RUNTIME-CONFIG-CLIENT-PRELOAD" class="id_link">#</a></h3></div></div></div><p>
      Several settings are available for preloading shared libraries into the
      server, in order to load additional functionality or achieve performance
      benefits.  For example, a setting of
      <code class="literal">'$libdir/mylib'</code> would cause
      <code class="literal">mylib.so</code> (or on some platforms,
      <code class="literal">mylib.sl</code>) to be preloaded from the installation's standard
      library directory.  The differences between the settings are when they
      take effect and what privileges are required to change them.
     </p><p>
      <span class="productname">PostgreSQL</span> procedural language libraries can
      be preloaded in this way, typically by using the
      syntax <code class="literal">'$libdir/plXXX'</code> where
      <code class="literal">XXX</code> is <code class="literal">pgsql</code>, <code class="literal">perl</code>,
      <code class="literal">tcl</code>, or <code class="literal">python</code>.
     </p><p>
      Only shared libraries specifically intended to be used with PostgreSQL
      can be loaded this way.  Every PostgreSQL-supported library has
      a <span class="quote">“<span class="quote">magic block</span>”</span> that is checked to guarantee compatibility.  For
      this reason, non-PostgreSQL libraries cannot be loaded in this way.  You
      might be able to use operating-system facilities such
      as <code class="envar">LD_PRELOAD</code> for that.
     </p><p>
      In general, refer to the documentation of a specific module for the
      recommended way to load that module.
     </p><div class="variablelist"><dl class="variablelist"><dt id="GUC-LOCAL-PRELOAD-LIBRARIES"><span class="term"><code class="varname">local_preload_libraries</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.4.6.1.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.4.6.1.1.4" class="indexterm"></a>
      </span> <a href="#GUC-LOCAL-PRELOAD-LIBRARIES" class="id_link">#</a></dt><dd><p>
        This variable specifies one or more shared libraries that are to be
        preloaded at connection start.
        It contains a comma-separated list of library names, where each name
        is interpreted as for the <a class="link" href="sql-load.html" title="LOAD"><code class="command">LOAD</code></a> command.
        Whitespace between entries is ignored; surround a library name with
        double quotes if you need to include whitespace or commas in the name.
        The parameter value only takes effect at the start of the connection.
        Subsequent changes have no effect.  If a specified library is not
        found, the connection attempt will fail.
       </p><p>
        This option can be set by any user.  Because of that, the libraries
        that can be loaded are restricted to those appearing in the
        <code class="filename">plugins</code> subdirectory of the installation's
        standard library directory.  (It is the database administrator's
        responsibility to ensure that only <span class="quote">“<span class="quote">safe</span>”</span> libraries
        are installed there.)  Entries in <code class="varname">local_preload_libraries</code>
        can specify this directory explicitly, for example
        <code class="literal">$libdir/plugins/mylib</code>, or just specify
        the library name — <code class="literal">mylib</code> would have
        the same effect as <code class="literal">$libdir/plugins/mylib</code>.
       </p><p>
        The intent of this feature is to allow unprivileged users to load
        debugging or performance-measurement libraries into specific sessions
        without requiring an explicit <code class="command">LOAD</code> command.  To that end,
        it would be typical to set this parameter using
        the <code class="envar">PGOPTIONS</code> environment variable on the client or by
        using
        <code class="command">ALTER ROLE SET</code>.
       </p><p>
        However, unless a module is specifically designed to be used in this way by
        non-superusers, this is usually not the right setting to use.  Look
        at <a class="xref" href="runtime-config-client.html#GUC-SESSION-PRELOAD-LIBRARIES">session_preload_libraries</a> instead.
       </p></dd><dt id="GUC-SESSION-PRELOAD-LIBRARIES"><span class="term"><code class="varname">session_preload_libraries</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.4.6.2.1.3" class="indexterm"></a>
      </span> <a href="#GUC-SESSION-PRELOAD-LIBRARIES" class="id_link">#</a></dt><dd><p>
        This variable specifies one or more shared libraries that are to be
        preloaded at connection start.
        It contains a comma-separated list of library names, where each name
        is interpreted as for the <a class="link" href="sql-load.html" title="LOAD"><code class="command">LOAD</code></a> command.
        Whitespace between entries is ignored; surround a library name with
        double quotes if you need to include whitespace or commas in the name.
        The parameter value only takes effect at the start of the connection.
        Subsequent changes have no effect.  If a specified library is not
        found, the connection attempt will fail.
        Only superusers and users with the appropriate <code class="literal">SET</code>
        privilege can change this setting.
       </p><p>
        The intent of this feature is to allow debugging or
        performance-measurement libraries to be loaded into specific sessions
        without an explicit
        <code class="command">LOAD</code> command being given.  For
        example, <a class="xref" href="auto-explain.html" title="F.3. auto_explain — log execution plans of slow queries">auto_explain</a> could be enabled for all
        sessions under a given user name by setting this parameter
        with <code class="command">ALTER ROLE SET</code>.  Also, this parameter can be changed
        without restarting the server (but changes only take effect when a new
        session is started), so it is easier to add new modules this way, even
        if they should apply to all sessions.
       </p><p>
        Unlike <a class="xref" href="runtime-config-client.html#GUC-SHARED-PRELOAD-LIBRARIES">shared_preload_libraries</a>, there is no large
        performance advantage to loading a library at session start rather than
        when it is first used.  There is some advantage, however, when
        connection pooling is used.
       </p></dd><dt id="GUC-SHARED-PRELOAD-LIBRARIES"><span class="term"><code class="varname">shared_preload_libraries</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.4.6.3.1.3" class="indexterm"></a>
      </span> <a href="#GUC-SHARED-PRELOAD-LIBRARIES" class="id_link">#</a></dt><dd><p>
        This variable specifies one or more shared libraries to be preloaded at
        server start.
        It contains a comma-separated list of library names, where each name
        is interpreted as for the <a class="link" href="sql-load.html" title="LOAD"><code class="command">LOAD</code></a> command.
        Whitespace between entries is ignored; surround a library name with
        double quotes if you need to include whitespace or commas in the name.
        This parameter can only be set at server start.  If a specified
        library is not found, the server will fail to start.
       </p><p>
        Some libraries need to perform certain operations that can only take
        place at postmaster start, such as allocating shared memory, reserving
        light-weight locks, or starting background workers.  Those libraries
        must be loaded at server start through this parameter.  See the
        documentation of each library for details.
       </p><p>
        Other libraries can also be preloaded.  By preloading a shared library,
        the library startup time is avoided when the library is first used.
        However, the time to start each new server process might increase
        slightly, even if that process never uses the library.  So this
        parameter is recommended only for libraries that will be used in most
        sessions.  Also, changing this parameter requires a server restart, so
        this is not the right setting to use for short-term debugging tasks,
        say.  Use <a class="xref" href="runtime-config-client.html#GUC-SESSION-PRELOAD-LIBRARIES">session_preload_libraries</a> for that
        instead.
       </p><div class="note"><h3 class="title">Note</h3><p>
        On Windows hosts, preloading a library at server start will not reduce
        the time required to start each new server process; each server process
        will re-load all preload libraries.  However, <code class="varname">shared_preload_libraries
        </code> is still useful on Windows hosts for libraries that need to
        perform operations at postmaster start time.
       </p></div></dd><dt id="GUC-JIT-PROVIDER"><span class="term"><code class="varname">jit_provider</code> (<code class="type">string</code>)
       <a id="id-1.6.6.14.4.6.4.1.3" class="indexterm"></a>
      </span> <a href="#GUC-JIT-PROVIDER" class="id_link">#</a></dt><dd><p>
        This variable is the name of the JIT provider library to be used
        (see <a class="xref" href="jit-extensibility.html#JIT-PLUGGABLE" title="30.4.2. Pluggable JIT Providers">Section 30.4.2</a>).
        The default is <code class="literal">llvmjit</code>.
        This parameter can only be set at server start.
       </p><p>
        If set to a non-existent library, <acronym class="acronym">JIT</acronym> will not be
        available, but no error will be raised. This allows JIT support to be
        installed separately from the main
        <span class="productname">PostgreSQL</span> package.
       </p></dd></dl></div></div><div class="sect2" id="RUNTIME-CONFIG-CLIENT-OTHER"><div class="titlepage"><div><div><h3 class="title">19.11.4. Other Defaults <a href="#RUNTIME-CONFIG-CLIENT-OTHER" class="id_link">#</a></h3></div></div></div><div class="variablelist"><dl class="variablelist"><dt id="GUC-DYNAMIC-LIBRARY-PATH"><span class="term"><code class="varname">dynamic_library_path</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.5.2.1.1.3" class="indexterm"></a>
      <a id="id-1.6.6.14.5.2.1.1.4" class="indexterm"></a>
      </span> <a href="#GUC-DYNAMIC-LIBRARY-PATH" class="id_link">#</a></dt><dd><p>
        If a dynamically loadable module needs to be opened and the
        file name specified in the <code class="command">CREATE FUNCTION</code> or
        <code class="command">LOAD</code> command
        does not have a directory component (i.e., the
        name does not contain a slash), the system will search this
        path for the required file.
       </p><p>
        The value for <code class="varname">dynamic_library_path</code> must be a
        list of absolute directory paths separated by colons (or semi-colons
        on Windows).  If a list element starts
        with the special string <code class="literal">$libdir</code>, the
        compiled-in <span class="productname">PostgreSQL</span> package
        library directory is substituted for <code class="literal">$libdir</code>; this
        is where the modules provided by the standard
        <span class="productname">PostgreSQL</span> distribution are installed.
        (Use <code class="literal">pg_config --pkglibdir</code> to find out the name of
        this directory.) For example:
</p><pre class="programlisting">
dynamic_library_path = '/usr/local/lib/postgresql:/home/my_project/lib:$libdir'
</pre><p>
        or, in a Windows environment:
</p><pre class="programlisting">
dynamic_library_path = 'C:\tools\postgresql;H:\my_project\lib;$libdir'
</pre><p>
       </p><p>
        The default value for this parameter is
        <code class="literal">'$libdir'</code>. If the value is set to an empty
        string, the automatic path search is turned off.
       </p><p>
        This parameter can be changed at run time by superusers and users
        with the appropriate <code class="literal">SET</code> privilege, but a
        setting done that way will only persist until the end of the
        client connection, so this method should be reserved for
        development purposes. The recommended way to set this parameter
        is in the <code class="filename">postgresql.conf</code> configuration
        file.
       </p></dd><dt id="GUC-EXTENSION-CONTROL-PATH"><span class="term"><code class="varname">extension_control_path</code> (<code class="type">string</code>)
      <a id="id-1.6.6.14.5.2.2.1.3" class="indexterm"></a>
      </span> <a href="#GUC-EXTENSION-CONTROL-PATH" class="id_link">#</a></dt><dd><p>
        A path to search for extensions, specifically extension control files
        (<code class="filename"><em class="replaceable"><code>name</code></em>.control</code>).  The
        remaining extension script and secondary control files are then loaded
        from the same directory where the primary control file was found.
        See <a class="xref" href="extend-extensions.html#EXTEND-EXTENSIONS-FILES" title="36.17.1. Extension Files">Section 36.17.1</a> for details.
       </p><p>
        The value for <code class="varname">extension_control_path</code> must be a
        list of absolute directory paths separated by colons (or semi-colons
        on Windows).  If a list element starts
        with the special string <code class="literal">$system</code>, the
        compiled-in <span class="productname">PostgreSQL</span> extension
        directory is substituted for <code class="literal">$system</code>; this
        is where the extensions provided by the standard
        <span class="productname">PostgreSQL</span> distribution are installed.
        (Use <code class="literal">pg_config --sharedir</code> to find out the name of
        this directory.) For example:
</p><pre class="programlisting">
extension_control_path = '/usr/local/share/postgresql:/home/my_project/share:$system'
</pre><p>
        or, in a Windows environment:
</p><pre class="programlisting">
extension_control_path = 'C:\tools\postgresql;H:\my_project\share;$system'
</pre><p>
        Note that the specified paths elements are expected to have a
        subdirectory <code class="filename">extension</code> which will contain the
        <code class="filename">.control</code> and <code class="filename">.sql</code> files; the
        <code class="filename">extension</code> suffix is automatically appended to
        each path element.
       </p><p>
        The default value for this parameter is
        <code class="literal">'$system'</code>. If the value is set to an empty
        string, the default <code class="literal">'$system'</code> is also assumed.
       </p><p>
        If extensions with equal names are present in multiple directories in
        the configured path, only the instance found first in the path will be
        used.
       </p><p>
        This parameter can be changed at run time by superusers and users
        with the appropriate <code class="literal">SET</code> privilege, but a
        setting done that way will only persist until the end of the
        client connection, so this method should be reserved for
        development purposes. The recommended way to set this parameter
        is in the <code class="filename">postgresql.conf</code> configuration
        file.
       </p><p>
        Note that if you set this parameter to be able to load extensions from
        nonstandard locations, you will most likely also need to set <a class="xref" href="runtime-config-client.html#GUC-DYNAMIC-LIBRARY-PATH">dynamic_library_path</a> to a correspondent location, for
        example,
</p><pre class="programlisting">
extension_control_path = '/usr/local/share/postgresql:$system'
dynamic_library_path = '/usr/local/lib/postgresql:$libdir'
</pre><p>
       </p></dd><dt id="GUC-GIN-FUZZY-SEARCH-LIMIT"><span class="term"><code class="varname">gin_fuzzy_search_limit</code> (<code class="type">integer</code>)
      <a id="id-1.6.6.14.5.2.3.1.3" class="indexterm"></a>
      </span> <a href="#GUC-GIN-FUZZY-SEARCH-LIMIT" class="id_link">#</a></dt><dd><p>
        Soft upper limit of the size of the set returned by GIN index scans. For more
        information see <a class="xref" href="gin.html#GIN-TIPS" title="65.4.5. GIN Tips and Tricks">Section 65.4.5</a>.
       </p></dd></dl></div></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="runtime-config-vacuum.html" title="19.10. Vacuuming">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="runtime-config.html" title="Chapter 19. Server Configuration">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="runtime-config-locks.html" title="19.12. Lock Management">Next</a></td></tr><tr><td width="40%" align="left" valign="top">19.10. Vacuuming </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 18.4 Documentation">Home</a></td><td width="40%" align="right" valign="top"> 19.12. Lock Management</td></tr></table></div></body></html>